INDEX

Explanations

references to physical locations or areas

oai_token-act-pair · gpt-3.5-turbo

references to various types of spaces in contexts of safety and accessibility

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bane

-0.76

razen

-0.70

 Browne

-0.67

od

-0.65

 OPEC

-0.63

oner

-0.63

 cartel

-0.62

hetamine

-0.61

iren

-0.61

mitter

-0.60

POSITIVE LOGITS

 spaces

3.92

 Spaces

2.59

 space

2.44

space

1.93

paces

1.68

 SPACE

1.68

 Space

1.56

 rooms

1.44

 spac

1.44

 environments

1.44

Activations Density 0.016%

references to physical locations or areas

references to various types of spaces in contexts of safety and accessibility

No Comments

No Known Activations